Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Agent Behavioral Safety on AgentDojo

97.1Safety Rate

Thought-Aligner-7B

51.65263.45175.2587.049May 16, 2025
Updated 7d ago

Evaluation Results

MethodLinks
2025.05
97.134.4
2025.05
96.838.5
2025.05
94.956.3
2025.05
9331.6
2025.05
92.945.7
2025.05
92.746.9
2025.05
9245.8
2025.05
90.351
2025.05
89.544.8
2025.05
88.547.9
2025.05
83.161.5
2025.05
74.362.5
2025.05
6764.6
2025.05
53.477.7